`cuda.core` 0.2.0 Release Notes#

Released on March 17, 2025

Highlights#

Add ProgramOptions to facilitate the passing of runtime compile options to Program.
Add pythonic access to Device and Kernel attributes.

The stream attribute is removed from LaunchConfig. Instead, the Stream object should now be directly passed to launch() as an argument.
The signature for launch() is changed by swapping positional arguments, the new signature is now (stream, config, kernel, *kernel_args)
Change __cuda_stream__ from attribute to method.
The Program.compile() method no longer accepts the options argument. Instead, you can optionally pass an instance of ProgramOptions to the constructor of Program.
Device.properties() now provides attribute getters instead of a dictionary interface.
The .handle attribute of various cuda.core objects now returns the underlying Python object instead of a (type-erased) Python integer.

Expose ObjectCode as a public API, which allows loading cubins from memory or disk. For loading other kinds of code types, please continue using Program.
A C++ helper function get_cuda_native_handle() is provided in the new include/utility.cuh header to retrive the underlying CUDA C objects (ex: CUstream) from a Python object returned by the .handle attribute (ex: Stream.handle).
For objects such as Program and Linker that could dispatch to different backends, a new .backend attribute is provided to query this information.
Support CUDA Event timing. (#481, #498, #508)
An Event may now be created without recording it to a Stream using the Device.create_event() method.
Program now supports the additional PTX code type. (#317)
Linker.link() exceptions now include the original error log. (#423)
In a systematic sweep through the cuda.core implementations, many exceptions messages were made more consistent and informative. (#458)

jit_lto_fractal.py — Demonstrates just-in-time link-time optimization for fractal generation. (Device, LaunchConfig, Linker, LinkerOptions, Program, ProgramOptions) (#475)
simple_multi_gpu_example.py — Example of using multiple GPUs. (Device, Program, LaunchConfig) (#304)
show_device_properties.py — Displays detailed device properties. (Device) (#474)

A dangling pointer problem in _linker.py was fixed. (#516)
Add @functools.lru_cache decorator for get_binding_version(). (#512)
Selected .decode() were changed to .decode("utf-8", errors="backslashreplace") to ensure that decoding error messages does not abort the process. (#510)
The performance of Device.compute_capability() was improved. (#459)
The Program constructor now issues a warning when falling back to cuLink(). (#315)
To avoid deprecation warnings, the cuda.bindings imports in the cuda.core implementations were cleaned up. (#404)